Traumatic brain injury (TBI) is a major cause of death and disability. Of the 3.5 million Americans who sustain a TBI every year, approximately 27,000 experience prolonged traumatic coma, the most severe form of TBI. Less than 20% of these patients make a good recovery, and most are left with life-long disabilities. ICU management of severe TBI focuses on monitoring intracranial pressure (ICP), but data from recently conducted randomized clinical trials indicate that this approach is overly simplistic. Another approach is to monitor the partial pressure of oxygen in brain tissue (PbtO2) and apply interventions to prevent brain tissue hypoxia and improve neurologic outcome. Clinical studies demonstrate that brain tissue hypoxia is common, that there is a strong relationship between low PbtO2 and poor outcome, and that timely interventions can reverse brain tissue hypoxia. The first randomized controlled trial of PbtO2 monitoring in severe TBI, titled ?Brain Oxygen Optimization in Severe TBI (BOOST) Phase 2,? enrolled 122 subjects and demonstrated that the mean hypoxia burden was reduced by 74% by the treatment protocol informed by PbtO2 monitoring (p < 0.0001), and there were no significant safety issues. There was a trend towards improved functional outcome, supporting the pre-determined non-futility hypothesis. We are proposing the BOOST-3 trial to determine if there is evidence of clinical efficacy of a treatment protocol based on PbtO2 monitoring compared to treatment based on ICP monitoring alone. BOOST-3 will enroll patients with severe TBI requiring placement of ICP monitors within 6 hours of presentation to a participating hospital. Patients will be randomized to a treatment protocol based on ICP monitoring alone or the combination of ICP and PbtO2 monitoring. The Glasgow Outcome Scale-Extended (GOS-E) measured at 6 months post injury will be the primary outcome. Other secondary outcomes include functional, cognitive and behavioral assessments at 6 months, safety, survival to discharge, shortened time to follow commands, and reduction of total brain hypoxia exposure.